Images as the source of information about life
Biologists must analyze traits in order to understand the significance of patterns in the two billion-year evolutionary history of life and to predict future effects of environmental change or genetic manipulation. Images are by far the most abundant source of documentation of life on the planet—but traits of organisms cannot be readily extracted from them.
The question: How do we take hundreds of thousands of images and use them to answer fundamental biological questions about ecology and evolution? At the very least, how do we extract traits, such as the example of a bird guide?
The answer: We make traits computable. Biology meets machine learning and vice versa.